Passive Service Checks and 127

[Chris Ditri]

Hello everyone.

This one is weird.  I am monitoring a cluster of co-located servers from our 
place of work.  We have to go through two firewalls to do this, so I decided 
to use passive service checks.  

Most of the time, Nagios is working great -- but every once in a while, (about 
once or twice a day) we get an error stating this: (Return code of 127 is out 
of bounds - plugin may be missing). 

I have seven machines I am monitoring with passive service checks, with an 
average of about eight processes.  The problem does not appear to be 
plugin-related, as this error does not seem to favor any one machine or 
service in the cluster.  I have each machine send out results on these checks 
every 5 minutes or so (each one staggered, to help ensure they don't all come 
in at once).  I have freshness_check_interval set to 665 -- a litte over 11 
minutes.

Now, if checks come in every 5 minutes, and checks do not go stale for 11 
minutes, theoretically I should never get this error (unless there is 
actually a problem).

At first, I thought it might be because I was using a beta version of nagios, 
but when 1.0 came out, I updated it and the problem still occurs.  Then I 
thought it might be the hub the nagios server was attached to, so we bought a 
new switch.  This, of course, did not help either.  I decided to log the 
output from nsca and it says that each check was sent successfully.

This makes no sense to me.  Prior to switching to Nagios, I used Netsaint with 
npre and nsca to acheive the same results and never had these problems -- and  
I had set my freshness checking interval to an even tighter window -- 7 
minutes.

Can anyone tell me what is going on here?

Thank  you!


Chris


-------------------------------------------------------
This SF.NET email is sponsored by:
SourceForge Enterprise Edition + IBM + LinuxWorld = Something 2 See!
http://www.vasoftware.com